Reducing unintended pregnancy in the United States.

JUST UNDER HALF OF ALL PREGNANCIES in the United States are unintended. There were more than 3 million such pregnancies in 2001, according to a recent study in Perspectives on Sexual and Reproductive Health. Seven years earlier, these statistics had been the same. One of every two women aged 15 to 44 in the United States has experienced at least one unintended pregnancy. What is responsible for the high incidence of unintended pregnancy, and what can be done to reduce it?

Information on levels of and trends in contraceptive use in the United States is found in the National Surveys of Family Growth. In these surveys, the National Center for Health Statistics interviews women ages 15 to 44--reproductive age--about childbearing, family planning and maternal and child health. Among the 61.6 million women in that age group in 2002, almost two-thirds were found to be using some method of contraception. Among the 38 percent who were not using any method, about one-fifth were at risk of unintended pregnancy; the rest had been sterilized for noncontraceptive reasons, were not having intercourse, were sterile, were pregnant or trying to become pregnant or had been pregnant less than two months earlier.

As can be seen in the table below, almost 90 percent of women at risk for unintended pregnancy were using a contraceptive method. Today, the most popular contraceptive methods are pills 11.6 million), female sterilization (10.3 million), male condoms (6.8 million) and male sterilization (3.5 million), according to National Center for Health Statistics data.

The mix of methods shown in the table resulted in a staggering 3.1 million unintended pregnancies in 2001. Forty-eight percent of unintended pregnancies result from contraceptive failure, the aforementioned Perspectives on Sexual and Reproductive Health study indicates, but 52 percent are contributed by the small minority of women at risk--10.7 percent in 2002, as the table shows--who use no method at all.

The next table shows estimates of contraceptive failure rates in the United States. Pregnancy rates during typical use show how effective different methods are during actual use, including inconsistent or incorrect use. Pregnancy rates during perfect use--following the directions for use, which for many methods entails doing so at every act of intercourse--show how effective methods can be. The difference between the two indicates how forgiving of imperfect use a method is.

Scrutiny of the table reveals four important findings: The most effective methods during typical use are generally those that do not require adherence, such as intrauterine devices and implants; methods requiring adherence, such as condoms, generally show a big difference between perfect-use and typical-use failure rates; even the least effective methods are much more effective than no method at all; and the most effective methods at preventing pregnancy do not protect against sexually transmitted infections.

Comparison of the two tables shows that intrauterine devices and implants, the most effective reversible long-term methods that do not require adherence, are not used very frequently. Implanon was not approved by the Food and Drug Administration until mid-2006 and thus could not appear in the first table here, but Norplant, which is no longer marketed, was used by only 1.3 percent of women at risk of pregnancy in 1995. This comparison suggests three strategies for reducing unintended pregnancy: increasing contraceptive use among those not using a method; increasing use of the most effective long-term methods not requiring adherence among those using less effective methods requiring adherence; and simultaneous use of condoms and more effective methods among those at risk of sexually transmitted infection.

Widespread use of emergency contraceptive pills as a last chance--after no method was used, or a condom broke or slipped off the penis, or several oral contraceptive pills were missed--could also help to reduce unintended pregnancy. What is the potential for increasing emergency contraception usage? In the past two decades, researchers have evaluated numerous interventions intended to accomplish this goal, including promotional campaigns, provision of emergency contraception pills in advance of need, distribution by pharmacists or over the counter and prescription by telephone. Most of these interventions have increased emergency contraception use, and some substantially so, but whether any of these increases were sizeable enough to produce a large public health effect is doubtful. Almost all studies that collected data on emergency contraception use found that only a minority of women exposed to the intervention ever used emergency contraception, and few of those who became pregnant had tried to prevent the pregnancy using emergency contraception. No study to date has directly shown that any intervention has actually reduced pregnancy rates. It seems unlikely that emergency contraception will ever have a dramatic public health impact.

The strategies suggested above, including widespread use of emergency contraceptive pills, would have the intended effect of reducing unintended pregnancy if they were successfully implemented, but family planning providers have long encouraged and promoted use of contraception, use of the most effective methods and use of emergency contraceptive pills, and yet the rate of unintended pregnancy remains high. Although advances in technology in the form of long-acting reversible methods not requiring adherence may be helpful, technology alone is not sufficient.

There has been considerable improvement in coverage of contraceptive supplies and services. Medicaid covers contraceptive supplies and services for poor women, and the federal government covers contraceptive supplies and services for its employees and their dependents. In 2002, almost every reversible contraceptive supply and service was covered by at least 89 percent of private insurance plans, and 86 percent of plans covered the five most popular prescription contraceptives, according to a 2004 Perspectives on Sexual and Reproductive Health article. Guttmacher Institute data indicate 24 states require private-sector insurers that cover prescription drugs to provide coverage for prescription contraceptives and related services (these mandates do not apply to the half of employees who have insurance through employers that self-insure). In those states, plans were much more likely to cover the five leading prescription contraceptives (87-92 percent versus 47-61 percent).

Women are still concerned about contraceptive costs, though. Some women are still not covered by insurance, and for many who are, co-pays and deductibles constitute a considerable economic burden. Coverage rules that impede consistent use also deserve attention. A delay in obtaining a needed contraceptive refill may result in significant reduction of contraceptive efficacy and is a very common reason for contraceptive failure, yet many insurance plans require women to fill prescriptions for contraception on a monthly basis. Because pregnancy is so expensive, marginal savings gained by limiting refill access are almost certain to be overwhelmed by added costs of pregnancy caused by reduced contraceptive effectiveness.

Reducing unintended pregnancy is a formidable challenge. There is no magic bullet. Between 1994 and 2001, the rate of unintended pregnancy declined among college graduates and the wealthiest women but increased among poor and less educated women. No matter how dedicated, family planning providers cannot fix such structural problems.

JAMES TRUSSELL and LISA WYNN are, respectively, director and associate research scholar at the Office of Population Research at Princeton University.

Percentage using among women at risk by age cohort

Contraceptive method            15-44    15-19    20-24    25-29

Pill                             27.2     43.5     46.1     33.7
Female sterilization             24.1      0.0      3.2     13.5
Condom                           16.0     22.1     20.2     18.4
No method                        10.7     18.0     12.1     10.5
Male sterilization                8.2      0.0      0.7      3.7
Depo-Provera                      4.8     11.4      8.8      5.8
Withdrawal                        3.6      2.1      4.5      6.9
Intrauterine device               1.9      0.2      1.6      3.3
Fertility awareness-based
  methods                         1.3      0.0      1.1      0.9
Calendar rhythm                   1.0      0.0      1.1      0.4
Implant, Lunelle or patch         1.2      1.0      1.3      2.2
Other methods (cervical cap,
  sponge and female condom)       0.6      1.0      0.1      0.4
Diaphragm                         0.3      0.0      0.1      0.4
Spermicides                       0.3      0.5      0.1      0.1

Number of women in cohort, percentage and number at risk

Number of women in millions      61.6      9.8      9.8      9.2
Percentage at risk               69.4     38.4     69.2     76.0
Number at risk in millions       42.7      3.8      6.8      7.0

Contraceptive method            30-34    35-39    40-44

Pill                             28.6     16.8     10.0
Female sterilization             24.9     37.2     45.8
Condom                           15.5     14.1     10.5
No method                         9.2      9.8      8.8
Male sterilization                8.4     12.8     16.8
Depo-Provera                      3.8      1.8      1.5
Withdrawal                        3.4      3.1      1.3
Intrauterine device               2.8      1.3      1.0
Fertility awareness-based
  methods                         1.5      1.8      2.0
Calendar rhythm                   1.2      1.4      1.5
Implant, Lunelle or patch         1.2      0.7      0.3
Other methods (cervical cap,
  sponge and female condom)       0.1      0.4      1.2
Diaphragm                         0.2      0.0      0.5
Spermicides                       0.4      0.3      0.3

Number of women in cohort, percentage and number at risk

Number of women in millions      10.3     10.9     11.5
Percentage at risk               76.2     78.6     75.9
Number at risk in millions        7.8      8.5      8.7

[Source: Contraceptive Technology, 2007, forthcoming.]

                                       Percentage of
                                     women experiencing
                                         unintended
                                      pregnancy within
                                       the first year      Percentage
                                           of use           of women
                                                           continuing
                                     Typical    Perfect      use at
Method                                 use        use       one year

No method                             85         85
Spermicides                           29         18            42
Withdrawal                            27          4            43
Fertility awareness-based methods     25                       51
  Standard Days method                            5
  TwoDay method                                   4
  Ovulation method                                3
Sponge
  Parous women                        32         20            46
  Nulliparous women                   16          9            57
Diaphragm with spermicide             16          6            57
Condom without spermicide
  Female (Reality)                    21          5            49
  Male                                15          2            53
Combined pill and minipill             8          0.3          68
Evra patch                             8          0.3          68
NuvaRing                               8          0.3          68
Depo-Provera                           3          0.3          56
IUD
  ParaGard (copper T)                  0.8        0.6          78
  Mirena (LNG-IUS)                     0.2        0.2          80
Implanon                               0.05       0.05         84
Female sterilization                   0.5        0.5         100
Male sterilization                     0.15       0.10        100

Emergency contraceptive pills: Treatment initiated within 72 hours
after unprotected intercourse reduces the risk of pregnancy by at
least 75 percent.

Lactational amenorrhea method: LAM is a highly effective temporary
method of contraception. To maintain effective protection against
pregnancy, another method must be used when menstruation resumes,
breastfeeding is reduced, bottle feeding is introduced or the baby
reaches age 6 months.

[Source: Contraceptive Technology, 2007, forthcoming.]